Chapter 4 Habitat Fragmentation and Interspecific Competition: Implications for Lynx Conservation Steven W. Buskirk, Department of Zoology and Physiology, University of Wyoming, P.O. Box 3166, Laramie, WY 82071 Leonard F. Ruggiero, Rocky Mountain Research Station, 800 E. Beckwith, P.O. Box 8089, Missoula, MT 59807 Charles J. Krebs, Department of Zoology, University of British Columbia, 6270 University Blvd., Vancouver, BC V6T 1Z4 Canada Abstract—Habitat fragmentation and interspecific competition are two important forces that potentially affect lynx populations. Fragmentation operates by various mechanisms, including direct habitat loss, vehicle collisions and behavioral distur- bance from roads, and changes in landscape features such as edges. Competition takes two forms: Exploitation competition involves potential competitors, such as coyotes and raptors, for food with lynx. Interference competition involves aggres- sive acts, almost always by a larger animal, that can include attacking and killing. Habitat fragmentation tends to facilitate competition by generalist predators, of which the most likely beneficiary is the coyote. Other potential interference competitors with lynx include cougars and bobcats. Of these three carnivores, all are more widespread and more abundant within the southern distribution of the lynx than 50 years ago. 83 Chapter 4—Buskirk Introduction Fragmentation (Dayan et al. 1989) and competition (Rosenzweig 1966) are major forces shaping the evolution, composition, and function of carnivore communities (“carnivorans” is a more precise term referring to members of the Order Carnivora; we use “carnivoran” and “carnivore” synonymously). In human-dominated landscapes, fragmentation and competition are strongly linked because vegetation mosaics in landscapes provide high quality environments for generalist species such as the coyote and great-horned owl (Goodrich and Buskirk 1995). Under such conditions, generalist predators are favored over habitat specialists such as the spotted owl and American marten. As humans change the patterns of natural landscapes through timber harvest, construction of roads and buildings, and conversion of land to other uses, ecological processes such as competition, dispersal, and predation are affected in various ways (Wilcove 1985). Although many of these effects are poorly understood, it is clear that the structure and function of animal communities can be dramatically altered (Wilcove et al. 1986; Yahner 1988; Oehler and Litvaitis 1996). Here, we discuss how interspecific competition affects carnivores in general, how habitat fragmentation may mediate this process, and how lynx populations in particular may be affected. Habitat Fragmentation The term “fragmentation” is used widely to describe human alterations of natural landscapes (Knight et al., in press). Lord and Norton (1990) defined this process as the disruption of continuity, especially as it relates to ecosystem processes. Forman (1995) discussed how fragmentation affects the area, size, shape, and configuration of landscape elements, in an overall process of land transformation that has major implications for conservation (Lord and Norton 1990; Wilcove et al. 1986). Fragmentation has been variously defined to describe a reduction of total area, increased isolation of patches, and reduced connectedness among patches of natural vegetation (Rolstad 1991). Fragmentation tends to reduce habitat area and to isolate patches of native vegetation (especially in late seral stages) from each other, both of which can lead to local species extirpations (Wilcox 1980; Wilcox and Murphy 1985). Moreover, the loss of some species in this way can lead to multiple extinctions through community-level secondary effects (Wilcox and Murphy 1985). Fragmentation has been applied to both natural (Andrén 1994) and human- caused alterations of landscape patterns. Here, we use the term “patchiness” 84 Buskirk—Chapter 4 when referring to natural processes (Buskirk et al., in press) and “fragmen- tation” to anthropogenic disruption of natural patterns and the secondary effects of such disruption (e.g., behavioral disturbance to wildlife), and the effects of cars, pets, garbage, and other human accoutrements (Buskirk et al., in press). Although appropriate to any landscape, fragmentation has been preferen- tially applied to forests (Simberloff 1994; Lehmkuhl and Ruggiero 1991; Harris 1984). In this context, some have argued that the negative effects of forest fragmentation are ameliorated in areas like the Rocky Mountains because of naturally patchy landscapes. This contention is unlikely to be true given that, by definition, fragmentation disrupts the natural landscape pattern regardless of the scale of the undisturbed pattern. That is, fragmen- tation is not scale-limited (Lord and Norton 1990). Moreover, natural land- scape pattern is a complex function of topography with its associated (e.g., edaphic, microclimatic) gradients and natural disturbance processes, the most important of which is fire. Clearly, the kinds, amounts, and arrange- ments of forested environments differ markedly between natural and anthropogenic disturbances (Harris 1984; Ruggiero et al. 1991). Fragmentation has been shown to affect a fairly wide range of birds and mammals (reviewed by Harris 1984, Bright 1993, Andrén 1994, and Oehler and Litvaitis 1996). This is not surprising, considering the wide range of mechanisms whereby vertebrates can be impacted—for example, loss of area, isolation of patches, vehicle collisions, increased predation by edge-preferring predators, changes in boundary conditions including altered moisture re- gimes near stand edges, and changes in the habitat matrix or context within which undisturbed habitats exist. Several of these mechanisms can mediate competitive relationships by permitting generalist predators such as coyotes access to landscapes occupied by specialist species that are ecologically separated in natural landscapes (Hunter 1990). Moreover, our consider- able understanding of ecological systems and mechanisms of competition provides a plausible basis for believing habitat-mediated competition is an important factor in population persistence. Finally, as clearly stated by Wilcox and Murphy (1985:884), “That current ecological theory is inad- equate for resolving many of the details should not detract from what is obvious and accepted by most ecologists: habitat fragmentation is the most serious threat to biological diversity and is the primary cause of the present extinction crisis.” Direct habitat effects of fragmentation of most concern in lynx conser- vation are (1) reduction of area and patch size of late-successional forest and of optimal snowshoe hare habitat; (2) creation of openings that facilitate access by potentially competing carnivores; (3) increased densities of edges 85 Chapter 4—Buskirk between early successional and other forest types; and (4) changes in the amounts and structural complexity of seral forest stands within landscapes. Although landscape-level studies have not determined how fragmentation affects lynx ecology and population persistence (Koehler and Aubry 1994), rare species associated with wilderness, such as the lynx, generally are considered most susceptible to fragmentation (Bright 1993). Likewise, habitat specialists with large individual spatial needs, including the lynx (Quinn and Parker 1987; O’Donoghue et al. 1998), are likely to be impacted by habitat fragmentation (Andrén 1994). This is so because generalist predators tend to dominate the predator guild in fragmented landscapes (Oehler and Litvaitis 1996). Competition Keddy (1989:2) defined competition as “the negative effects which one organism has upon another by consuming, or controlling access to, a resource that is limited in availability.” Two qualitatively different kinds of interspecific competition can be illustrated by examples involving the lynx (Fig. 4.1). Exploitation (resource) competition (Litvaitis 1992) occurs when other species, such as the northern goshawk, bobcat, or coyote, use resources that limit the fitness of a lynx. Thus, if northern goshawks exploit snowshoe hares in an area so thoroughly that lynx die sooner (from starvation or while dispersing to a new place), breed less (from females failing to mate or give birth), or produce smaller litters of kittens than they would otherwise, we say that northern goshawks competed (exploitatively) with lynx. Although the competition might have been reciprocal, that issue is academic from our perspective here. Interference competition (Case and Gilpin 1974) occurs when one species acts aggressively toward another, denying it access to a resource (Fig. 4.1). For example, if a cougar were to chase a lynx away from a hare carcass, or kill one of its kittens, or scent mark an area so that lynx were deterred from foraging there, this would constitute interference competition. Exploitation competition is not particularly affected by the relative body sizes of the participants, as in the hypothetical example of the smaller goshawk and the larger-bodied lynx. Interference competition, by contrast, is almost invari- ably inflicted by a larger carnivore on a smaller one (Fig. 4.2). Further, the likelihood of interference competition among carnivores seems to be great- est when two species are similar in body form and size (Buskirk in press); for example, wolves are more likely to exert interference competition on coyotes than on red foxes (similar shape but different size: Johnson